TY - JOUR
AU - Steinhäuser, Christian
AB - Abstract The ventral posterior nucleus of the thalamus plays an important role in somatosensory information processing. It contains elongated cellular domains called barreloids, which are the structural basis for the somatotopic organization of vibrissae representation. So far, the organization of glial networks in these barreloid structures and its modulation by neuronal activity has not been studied. We have developed a method to visualize thalamic barreloid fields in acute slices. Combining electrophysiology, immunohistochemistry, and electroporation in transgenic mice with cell type-specific fluorescence labeling, we provide the first structure-function analyses of barreloidal glial gap junction networks. We observed coupled networks, which comprised both astrocytes and oligodendrocytes. The spread of tracers or a fluorescent glucose derivative through these networks was dependent on neuronal activity and limited by the barreloid borders, which were formed by uncoupled or weakly coupled oligodendrocytes. Neuronal somata were distributed homogeneously across barreloid fields with their processes running in parallel to the barreloid borders. Many astrocytes and oligodendrocytes were not part of the panglial networks. Thus, oligodendrocytes are the cellular elements limiting the communicating panglial network to a single barreloid, which might be important to ensure proper metabolic support to active neurons located within a particular vibrissae signaling pathway. astrocyte, barreloid, gap junction coupling, oligodendrocyte, thalamus Introduction The thalamus plays an important role in relay and modulation of information to the neocortex. The ventral posterior medial nucleus (VPM) of the ventrobasal thalamus is part of the somatosensory system. It contains an array of elongated cellular domains called barreloids, which provide a structural basis for the somatotopic organization of vibrissae representation (Van Der 1976; Land et al. 1995). As part of the whisker pathway, barreloids receive sensory input from individual vibrissae and transmit their output to the corresponding cortical barrels (Sugitani et al. 1990; Mosconi et al. 2010). It is increasingly appreciated that astrocytes represent a morphologically and functionally heterogeneous cell population, which form extended gap junction-coupled networks (Theis and Giaume 2012). Gap junction channels are composed of connexins (Cx) and the different isoforms are expressed in a cell type-specific manner (Bedner et al. 2012). Functional coupling among astrocytes is mainly based on Cx43 (gene name, Gja1) and Cx30 (Gjb6) while gap junctions between oligodendrocytes are formed by Cx47 (Gjc2) and Cx32 (Gjb1). Recent studies have shown that astrocytes in the thalamus differ in various aspects from their counterparts in other brain regions. For instance, thalamic astrocytes are heterogeneous in expressing AMPA-type glutamate receptors (Höft et al. 2014), and extensive panglial coupling between astrocytes and oligodendrocytes is a prominent pathway of inter-glial communication in the thalamus but less so in the hippocampus or neocortex (Griemsmann et al. 2015). Astrocytic gap junction networks in the hippocampus are critical for the delivery of energetic metabolites to maintain synaptic activity (Rouach et al. 2008). Conversely, neuronal activity may regulate the permeability of astrocytic gap junction channels, as has been shown in the olfactory bulb (Roux et al. 2011). Moreover, previous work demonstrated that the shape of astrocyte coupling networks may be restricted to and oriented along neuronal compartments. Such confined networks are found in the barrel fields of the neocortex, where intercellular astroglial coupling preferentially occurs within the barrels, while coupling across the borders of these columns is virtually absent (Houades et al. 2008). Compared with barrel fields in the neocortex, much less is known about the cellular composition and properties of thalamic barreloids. Notably, not a single study, so far has addressed the functional properties of glial cells and their impact on neuronal signaling in these domains; the standard work “The thalamus” (Jones, 2007) even lacks the key words glia, astrocyte, or oligodendrocyte in the index. In this study, we developed a method to visualize thalamic barreloids in acute brain slices, which allowed for structure-function analyses. Our findings reveal abundant intercellular communication between astrocytes and oligodendrocytes, but at the same time also demonstrate functional heterogeneity among barreloidal glial cells, indicating that they might have distinct roles in controlling neural microcircuits. Materials and Methods Experiments were performed with tissue from transgenic mice with astrocyte- (human glial fibrillary acidic protein-enhanced green fluorescence (hGFAP-EGFP); Nolte et al. 2001) or oligodendrocyte-specific fluorescence labeling (myelin proteolipid protein-green fluorescence protein (PLP-GFP); Fuss et al. 2000) and from wild-type mice (C57/Bl6J/N, Charles River) of either sex, aged between postnatal days (p) 14–17. Mice were kept under standard housing conditions. All experiments were carried out in accordance with local, state, and European regulations. Electrophysiology Mice were anaesthetized with isoflurane (Abbott) and decapitated. Brains were removed and cut in 200 μm thick slices using a vibratome (VT1200S, Leica) filled with ice-cold preparation solution (1.25 mM NaH2PO4, 87 mM NaCl, 2.5 mM KCl, 7 mM MgCl2, 0.5 mM CaCl2, 25 mM glucose, 25 mM NaHCO3, 75 mM sucrose) bubbled with carbogen (95% O2, 5% CO2) at pH 7.4. The cutting plane was optimized to allow visualization of the barreloids in acute slices (see Results). The slices were transferred to carbogenized preparation solution (35°C, 20 min) and stored in carbogenized artificial cerebrospinal fluid (ACSF; 1.25 mM NaH2PO4 (Sigma Aldrich), 126 mM NaCl, 3 mM KCl, 2 mM MgSO4, 2 mM CaCl2, 10 mM glucose, 26 mM NaHCO3) at room temperature (RT; pH 7.4). Slices were then transferred to a recording chamber and continuously perfused with carbogenized ACSF. Barreloids and cells were visualized using a microscope (Axioskop, Zeiss) equipped with a CCD camera (PCO) using 10× (Zeiss) and 60× (Olympus) objectives. Pipettes were fabricated from borosilicate glass capillaries with an outer diameter of 2 mm (Hilgenberg) and had resistances of 3–5 MΩ when filled with the internal solution (130 mM K-gluconate, 1 mM MgCl2, 3 mM Na2-ATP, 20 mM 10 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES), 10 mM ethyleneglycol-bis(2-aminoethylether)-N,N,N′,N′-tetra acetic acid (EGTA)), pH 7.2, supplemented with 0.5% N-biotinyl-l-lysine (biocytin, Sigma Aldrich) and 0.1% Texas Red Dextran (3 kDa, Invitrogen). For glucose diffusion experiments, the internal solution (105 mM K-gluconate, 1 mM MgCl2, 3 mM Na2-ATP, 10 mM HEPES, 5 mM EGTA, 30 mM KCl, 0.5 mM CaCl2, pH 7.4, 285 mOsm) contained the glucose analog 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]−2-deoxyglucose (2-NBDG, Molecular Probes, Life Technologies) (2 mg/mL) supplemented with 0.5% biocytin and 0.1% Texas Red Dextran. The tracer biocytin (372 Da) permeates gap junctions while Texas Red Dextran (3 kDa) is impermeable and hence was confined to the initially filled cell. Currents were recorded at RT using an EPC-800 or EPC-9 patch clamp amplifier (Heka) and monitored by TIDA software (Heka). Thalamic neurons fired spontaneous action potentials as revealed with current clamp recordings (initial resting potential −63.5 ± 2.7 mV, 5 cells, corrected for a liquid junction potential of −7 mV; not shown). In slices from hGFAP-EGFP mice, astrocytes were identified according to their bright green fluorescence and their morphology (Matthias et al. 2003). These cells displayed electrophysiological properties typical of astrocytes (series resistance (RS) = 7.6 ± 0.1 MΩ; membrane resistance (RM) = 2.1 ± 0.1 MΩ, resting potential −77 ± 1.4 mV; 38 cells). To identify astrocytes in thalamic slices from PLP-GFP mice, the tissue was incubated in ACSF supplemented with sulforhodamine 101 (SR101; 1 µM, 35°C, 20 min; Molecular Probes) (Griemsmann et al. 2015). For gap junction coupling analysis, astrocytes were filled with internal solution for 20 min in the voltage clamp mode. The holding potential was −80 mV and currents were recorded every 10 min. To study the modulation of coupling by neuronal activity, slices were incubated with tetrodotoxin (TTX) (Abcam, ab120055) and ω-conotoxin GVIA (Sigma Aldrich) prior to (for 3–4 h) and during the recording. After recording, we fixed the slices in 4% paraformaldehyde (PFA) in phosphate buffered saline (PBS), pH 7.4 (4°C, overnight), and stored them in PBS. If not stated, otherwise, chemicals were purchased from AppliChem. Visualization of Tracer Coupling For visualization of biocytin and barreloid structures, after recording slices were washed (3× in PBS) and incubated with blocking solution (10% normal goat serum (NGS), Merck Millipore, and 2% Triton X-100 in PBS; 2 h, RT). Then they were incubated with streptavidin-Cy5 (1:200; Dianova; S32357) or streptavidin-Alexa Fluor 647 (1:600; Molecular Probes, 016–170-084) in PBS with NGS (2%) and Triton X-100 (0.1%) at 4°C, overnight. The next day, slices were washed in PBS and stained with Hoechst (Invitrogen) (1:100 in dH2O, 10 min, RT). After washing they were mounted using Aquapolymount (Polysciences Europe). Slices from PLP-GFP mice were blocked for 4 h (10% NGS, 5% Triton X-100, in PBS) and stained for GFP using chicken anti-GFP (Abcam, ab13970) (1:500, 4°C, overnight) and goat anti-chicken Alexa Fluor 488 (Invitrogen, A11039) (1:500, 2% NGS, 0.1% Triton X-100, in PBS, 2 h, RT). Single-Cell Electroporation Single-cell electroporation experiments (Nevian and Helmchen 2007) were performed using a TCS SP5 microscope (Leica). Slices were perfused with ACSF at RT. Pipettes fabricated as described above were filled with 1 mM Alexa Fluor 594 in dH2O. Neurons were visually identified within barreloids and filled with Alexa Fluor 594. For that, a pipette was positioned close to the membrane (distance ~1 µm) and a single voltage pulse was applied (−10 to −12 V, 20–30 ms). Thereby a transient pore in the plasma membrane was generated and dye diffused into the soma. Alexa Fluor 594 fluorescence of 8–12 electroporated neurons was analyzed using 2-photon excitation fluorescence imaging. Stacks of optical sections were obtained with a Leica TCS SP5 microscope equipped with an infrared ultra-short-pulse laser (MaiTai; Spectra Physics). Two-photon absorption was achieved by excitation of the fluorophores with femtosecond pulses of infrared light with repetition rates of 80 MHz. The wavelength for excitation was adjusted for best signal to noise ratio for Alexa Fluor 594 and set to 810 nm at approximately 2 W at the output port. Fluorescence was collected with a nondescanned detector. Volumes of 337 × 337 × 80–90 μm3 (x/y/z) were scanned with a resolution of 1024 × 1024 × 80–90 pixels. Image acquisition recording was performed with Leica LAS AF software and analyzed with Imaris 8.1 (Bitplane AG). Cytochrome Oxidase Staining Mice were anesthetized with isoflurane and decapitated. Brains were removed, fixed in PFA (4% in PBS, pH 7.4, 4°C, overnight) and stored in PBS. Sections (40 μm) were cut as described in Results (Fig. 1) using a vibratome (VT1200S) and incubated with a staining medium (0.2 mM DAB, Sigma Aldrich, 100 mL PBS, pH 7.4, 4.03 mM cytochrome C, Sigma Aldrich, 11.4 mM sucrose, 3–6 h, 37°C (Wong-Riley 1979)). After 3 times washing in PBS, they were mounted using a gelatine-glycerol medium (5 g gelatine, Sigma Aldrich, 50 mL distilled water, 50 mL glycerol, Sigma Aldrich, pH 7.0). Figure 1. View largeDownload slide An oblique cutting angle visualizes thalamic barreloids in acute brain slices. Visualization of barreloids in acute brain slices of juvenile mice was achieved by modifying the horizontal slice preparation. (A) Scheme of the cutting plane. The inset (top) gives a mouse brain in top view indicating rostral (r), lateral (l), and caudal (c) directions. The blue area is also seen in the larger scheme (d, dorsal; v, ventral; m, medial). The cutting plane (indicated in red) is tilted up 5° anteriorly and 30° laterally from the horizontal plane. This cutting plane allowed visualizing the elongated barreloid fields in acute slices with differential interference contrast optics (inset bottom; barreloid borders appear dark, a patch pipette is seen above tissue, scale bar 100 µm). (B) At p14, cytochrome oxidase staining of fixed slices obtained from hGFAP-EGFP mice visualized barreloids in the thalamus (box; TH, thalamus) and the barrels in the somatosensory cortex (BC, barrel cortex; HC, hippocampus). Scale bar, 500 µm. (C) Higher magnification of the boxed area in (B). The dark, elongated structures represent barreloids, light structures demarcate barreloid borders. Scale bar, 250 µm. VPM, ventral posterior medial nucleus. Figure 1. View largeDownload slide An oblique cutting angle visualizes thalamic barreloids in acute brain slices. Visualization of barreloids in acute brain slices of juvenile mice was achieved by modifying the horizontal slice preparation. (A) Scheme of the cutting plane. The inset (top) gives a mouse brain in top view indicating rostral (r), lateral (l), and caudal (c) directions. The blue area is also seen in the larger scheme (d, dorsal; v, ventral; m, medial). The cutting plane (indicated in red) is tilted up 5° anteriorly and 30° laterally from the horizontal plane. This cutting plane allowed visualizing the elongated barreloid fields in acute slices with differential interference contrast optics (inset bottom; barreloid borders appear dark, a patch pipette is seen above tissue, scale bar 100 µm). (B) At p14, cytochrome oxidase staining of fixed slices obtained from hGFAP-EGFP mice visualized barreloids in the thalamus (box; TH, thalamus) and the barrels in the somatosensory cortex (BC, barrel cortex; HC, hippocampus). Scale bar, 500 µm. (C) Higher magnification of the boxed area in (B). The dark, elongated structures represent barreloids, light structures demarcate barreloid borders. Scale bar, 250 µm. VPM, ventral posterior medial nucleus. Immunohistochemistry Mice were anesthetized by intraperitoneal injection of 80 mg/kg ketamine hydrochloride (Medistar) and 1.2 mg/kg medetomidine hydrochloride (CP Pharma) and perfused with PBS followed by 4% PFA. Dissected brains were postfixed in 4% PFA in PBS, pH 7.4 at 4°C overnight, and transferred in 30% sucrose in PBS for 2–3 days. Frozen brains were cut (80 μm thick) as described in Results using a vibratome (VT1200S). Slices were washed (3× in PBS) and incubated with blocking solution (10% NGS, 0.4% Triton X-100, in PBS, 2 h, RT). Then they were incubated with the primary and secondary antibody in 0.1% Triton X-100, 2% NGS, in PBS overnight. The following primary antibodies were used: chicken anti-GFP (1:400, Abcam, ab13970), mouse anti-neurofilament SMI312 (1:200, Covance; SMI-312 R), and mouse anti-NeuN (1:200, Chemicon; MAB377). The next day slices were washed (3× in PBS) and incubated with the corresponding secondary antibodies (1.5 h, RT): goat anti-mouse Alexa Fluor 594 (1:500, Invitrogen, A11032), goat anti-rabbit Alexa Fluor 594 (1:500, Invitrogen, A11037), goat anti-chicken Alexa Fluor 488 (1:500, Invitrogen, A11039), and goat anti-mouse Alexa Fluor 647 (1:500, Invitrogen, A21235). After washing (3× in PBS), slices were stained with Hoechst (1:100 in dH2O, 10 min, RT, Invitrogen), washed and mounted using Aquapolymount (Polysciences). Data Analysis For coupling analyses, x/y/z-stack images (z-spacing 1–2 µm) were obtained with a confocal microscope (TCS SP8 Leica) or a microscope equipped with epifluorescence (Axiophot, Zeiss) employing MetaVue software (Molecular Devices) using 5×, 20×, and 40× objectives (Zeiss). Two different approaches were used for coupling analysis. To capture strength and anisotropy of gap junction coupling between cells in hGFAP-EGFP mice (Fig. 2), the number of coupled cells and the shape of the cloud of visualized biocytin were analyzed. To this end, the z-plane containing the cell that was initially filled with biocytin (i.e., the cell with Texas Red fluorescence) was identified. We then counted the biocytin-positive cells in this plane only. The spatial extent of coupling in parallel and orthogonal to barreloid borders was analyzed by obtaining 2 orthogonal fluorescence line profiles through the center of the dye cloud (Fig. 2A2 and C1 for an illustration, length 300 µm in 4 µm bins, width 100 µm), which were normalized to the fluorescence intensity at the patched cell. The width of the fluorescence profile at half maximum was measured. The ratio of widths (parallel over orthogonal) was used as a measure of the anisotropy of coupling. The displacement of the center of the coupled network away from the filled cell was used as an indicator for the presence of a barrier for dye/tracer diffusion (Anders et al. 2014). This was defined by the distance between the calculated network center and the position of the filled cell in the fluorescence profile orthogonal to the barreloid border. Figure 2. View largeDownload slide Barreloid borders restrict dye coupling in thalamic glial networks. Acute brain slices containing barreloids were prepared from hGFAP-EGFP mice (p14–17). One astrocyte per slice was identified by its intrinsic fluorescence, typical morphology, and passive current pattern upon de- and hyperpolarization of the membrane between −160 and +20 mV (10 mV increment, holding potential −80 mV; scale bars, 12.5 ms, 5 nA; insets). During recording, the cell was filled with biocytin (20 min) included in the patch pipette solution. (A) An astrocyte in the center of a barreloid field was filled. The low magnification image (A1) shows a merge of the coupled biocytin-filled network (red) and Hoechst staining (white). The short white lines indicate the barreloid borders. (A2) gives at higher resolution a confocal image (one optical plane) of the same network. The white arrow head indicates the initially filled astrocyte. Note the oval shape of this network. (B) Filling a thalamic astrocyte outside the barreloid fields gave rise to a spherical coupling network, with the initially filled cell (arrow head) being located in its center (B1, B2). (C1) The shape of the coupling networks was assessed by calculating normalized fluorescence profiles along orthogonal lines in parallel (red) and orthogonal (orange) to the barreloid borders (position and size of fluorescence line profiles illustrated in A2). (C2) The ratio of parallel and perpendicular widths of the fluorescence profiles (full width at half maximum) was above one and significantly higher within barreloid fields (gray, 19 slices from 13 mice) than in the spherically shaped thalamic networks outside the fields (black, 19 slices from 12 mice). (D) Filling an astrocyte within a barreloid close to its border (D1, plotted gray lines indicate barreloid borders) resulted in asymmetric diffusion, that is, the initially filled cell (arrow head) was apart from the visualized network center (D2). (E) The distance of the initially filled cell from the calculated coupling network center (see Materials and Methods) was significantly larger when cells close to the barreloid border were filled (br-border, 11 slices from 9 mice) compared with cells filled in the barreloid center (br-center, 8 slices from 6 mice) or outside barreloid fields (black, 19 slices from 12 mice). Scale bars, 100 µm (A1, B1, D1) and 50 µm (A2, B2, D2). Number of mice is given in bar graphs. Figure 2. View largeDownload slide Barreloid borders restrict dye coupling in thalamic glial networks. Acute brain slices containing barreloids were prepared from hGFAP-EGFP mice (p14–17). One astrocyte per slice was identified by its intrinsic fluorescence, typical morphology, and passive current pattern upon de- and hyperpolarization of the membrane between −160 and +20 mV (10 mV increment, holding potential −80 mV; scale bars, 12.5 ms, 5 nA; insets). During recording, the cell was filled with biocytin (20 min) included in the patch pipette solution. (A) An astrocyte in the center of a barreloid field was filled. The low magnification image (A1) shows a merge of the coupled biocytin-filled network (red) and Hoechst staining (white). The short white lines indicate the barreloid borders. (A2) gives at higher resolution a confocal image (one optical plane) of the same network. The white arrow head indicates the initially filled astrocyte. Note the oval shape of this network. (B) Filling a thalamic astrocyte outside the barreloid fields gave rise to a spherical coupling network, with the initially filled cell (arrow head) being located in its center (B1, B2). (C1) The shape of the coupling networks was assessed by calculating normalized fluorescence profiles along orthogonal lines in parallel (red) and orthogonal (orange) to the barreloid borders (position and size of fluorescence line profiles illustrated in A2). (C2) The ratio of parallel and perpendicular widths of the fluorescence profiles (full width at half maximum) was above one and significantly higher within barreloid fields (gray, 19 slices from 13 mice) than in the spherically shaped thalamic networks outside the fields (black, 19 slices from 12 mice). (D) Filling an astrocyte within a barreloid close to its border (D1, plotted gray lines indicate barreloid borders) resulted in asymmetric diffusion, that is, the initially filled cell (arrow head) was apart from the visualized network center (D2). (E) The distance of the initially filled cell from the calculated coupling network center (see Materials and Methods) was significantly larger when cells close to the barreloid border were filled (br-border, 11 slices from 9 mice) compared with cells filled in the barreloid center (br-center, 8 slices from 6 mice) or outside barreloid fields (black, 19 slices from 12 mice). Scale bars, 100 µm (A1, B1, D1) and 50 µm (A2, B2, D2). Number of mice is given in bar graphs. Panglial coupling in PLP-GFP mice (Fig. 3) was quantified by counting the total number of biocytin-filled cells across all z-plane of an image stack. This 3D counting was also used to estimate the spread of 2-NBDG in C57Bl/6 and PLP-GFP mice and its dependence on neuronal activity. To this end, intercellular 2-NBDG fluorescence was captured online with 2-photon excitation fluorescence imaging (TCS SP5, Leica) 20 min after loading and analyzed offline with Fiji software. Figure 3. View largeDownload slide Panglial coupled networks in barreloids contain many oligodendrocytes. Astrocytes were identified through SR101 labeling in the thalamus of juvenile PLP-GFP mice. One astrocyte per slice was filled with biocytin and coupling was visualized as described in the legend to Fig. 2. (A) Networks within the barreloid fields (A1) were predominantly composed of GFP-positive oligodendrocytes (A2, A3). Note that many of the GFP-positive cells were located directly at the barreloid borders. (B) Thalamic coupling networks outside the barreloids (B1) also contained many GFP-positive oligodendrocytes (B2, B3), which, however, were more evenly distributed. Scale bar, 50 µm. Note that several PLP-GFP-positive cells within the coupled domain lack biocytin. (C) Networks within the barreloid fields (7 slices from 7 mice) contained more PLP-GFP-positive cells than those outside (15 slices from 11 mice). Number of mice is given in bar graphs. Figure 3. View largeDownload slide Panglial coupled networks in barreloids contain many oligodendrocytes. Astrocytes were identified through SR101 labeling in the thalamus of juvenile PLP-GFP mice. One astrocyte per slice was filled with biocytin and coupling was visualized as described in the legend to Fig. 2. (A) Networks within the barreloid fields (A1) were predominantly composed of GFP-positive oligodendrocytes (A2, A3). Note that many of the GFP-positive cells were located directly at the barreloid borders. (B) Thalamic coupling networks outside the barreloids (B1) also contained many GFP-positive oligodendrocytes (B2, B3), which, however, were more evenly distributed. Scale bar, 50 µm. Note that several PLP-GFP-positive cells within the coupled domain lack biocytin. (C) Networks within the barreloid fields (7 slices from 7 mice) contained more PLP-GFP-positive cells than those outside (15 slices from 11 mice). Number of mice is given in bar graphs. To analyze the orientation of neuronal processes relative to barreloid borders, image stacks of groups of electoporated neurons were obtained. Imaris 8.2 (Bitplane) was used to generate 3D reconstructions and to compute iso-surfaces. In addition to fluorescence intensity, a criterion for data inclusion was a volume of at least 3000 voxels. Each calculated volume fragment was fitted into an ellipsoid. Since the 3D distribution of angles of spatial ellipsoid orientation did not follow a normal distribution (Shapiro–Wilk test), we used the median of all ellipsoid principal axes as a sum vector to indicate neuronal orientation. The upper and lower borders of the barreloids from corresponding brightfield pictures were considered as vectors and averaged to a sum vector. The angle between the 2 sum vectors, neuronal orientation, and barreloid orientation was determined for each brain slice. The result of this analysis was given as median and range, since the angle distributions (sum vectors of borders vs. ellipsoids) were not normally distributed (Shapiro–Wilk test). With the exception of neuronal orientation, all results are given as mean ± standard error of the mean (SEM) and n refers to the number of mice investigated. The number of slices per condition is given in the figure legends or in Results. Data were tested with Student's t-test or analysis of variance followed by Tukey's test as appropriate. Differences were regarded as significant at *P < 0.05  or **P < 0.01. Results Visualization of Barreloid Structures in Acute Brain Slices To visualize barreloid structures of juvenile mice in acute brain slices for coupling analysis, the cutting plane through the VPM had to be optimized. The barreloids were best visible when tissue blocks were tilted up 30° laterally and 5° anteriorly from the horizontal plane during preparation of acute slices (Fig. 1A). To confirm that the visualized structures in acute slices corresponded to thalamic barreloids, we performed cytochrome oxidase staining on fixed slices, which is known to tinge barrels in the somatosensory cortex as well as barreloids (Fig. 1B,C) (Mosconi et al. 2010). After fixation, barreloid borders were no longer visible with brightfield illumination. Different staining techniques were tested to allow visualization of barreloid borders and biocytin-filled networks within the same slices. We found that Hoechst staining of nuclei was best suited to simultaneously illustrate barreloid borders and coupled networks. Barreloid Borders Limit Biocytin Diffusion Comparative biocytin coupling analyses were performed in the VPM of juvenile hGFAP-EGFP mice, within and outside the barreloid structures by filling individual astrocytes with biocytin for 20 min. Astrocytes were identified based on their intrinsic fluorescence, characteristic morphology, and passive current pattern (Griemsmann et al. 2015). Barreloids in juvenile mice had a width of 50–120 µm and a length 700 µm. The number of coupled cells in the plane of the initially filled astrocyte was similar within barreloids (25.8 ± 1.7, 19 slices from 13 mice) and outside the barreloid field (37.0 ± 5.9 cells, 19 slices from 12 mice). Visual inspection of the biocytin-filled networks revealed that networks within barreloids were oval-shaped, whereas networks outside were circular. To determine the shape of the networks, we analyzed fluorescence line profiles (cf. Fig. 2A2, C1). The full widths at half-maximal intensity were calculated in parallel and orthogonal to barreloid borders and the ratio was taken. By definition, this ratio is 1 for spherical and >1 for elongated networks. Coupling networks within barreloids displayed ratios above unity (1.53 ± 0.08, n = 19 slices from 13 mice). In contrast, coupling networks outside of barreloids had a significantly lower ratio, close to unity (0.95 ± 0.04, 19 slices from 12 mice). Thus, biocytin diffusion between the coupled glial cells within barreloids is anisotropic and occurs preferentially within and along barreloids. This suggests that barreloid borders form a barrier for diffusion between glia cells. We have previously established how an anatomical border can modify dye coupling within a network and established the center of mass displacement as a useful parameter (Anders et al. 2014). In short, if a tracer is injected into a cell close to an anatomical border the tracer will preferentially diffuse away from it such that the center of mass of tracer is different from the site of injection. As illustrated in Fig. 2D, such asymmetric diffusion of the tracer was observed, that is, the visualized network center is no longer located at the initially filled astrocyte, when an astrocyte close to a barreloid border was filled with biocytin. Using a simplified method of quantification (see Materials and Methods), we found that the displacement is significantly higher when cells close to the barreloid border were filled (displacement outside barreloids: 7.1 ± 1.0 µm, 19 slices from 12 mice; displacement barreloid center: 6.2 ± 1.3 µm, 8 slices from 6 mice; displacement near barreloid border: 13.7 ± 2.3 µm, 11 slices from 9 mice, Fig. 2E). Together these data reveal that barreloid borders form a barrier for a coupled glial network thus restricting glial gap junction signaling to an individual barreloid field. Coupling Networks Within Barreloids Comprise Astrocytes and Oligodendrocytes We have previously reported that coupling networks in the thalamus of adult mice consist of astrocytes and oligodendrocytes while panglial coupling is much less abundant in the hippocampus (Griemsmann et al. 2015). To test whether oligodendrocytes also contribute to the coupling networks within barreloids of juvenile animals, PLP-GFP mice were used. Astrocytes were identified by incubating the tissue with SR101 and then were filled with biocytin during patch clamp recording. Similar to the adult thalamus (Griemsmann et al. 2015), outside the barreloids about half the coupled cells were PLP-GFP-positive (51.9 ± 2.9%, 15 slices from 11 mice), while within these structures the proportion of coupled oligodendrocytes was even higher (64.0 ± 5.0%, 7 slices from 7 mice, P = 0.03) (Fig. 3). The number of coupled cells outside the barreloids was higher (155.8 ± 18.1 cells, 15 slices from 11 mice) than in barreloids (96.6 ± 10.1 cells, 7 slices from 7 mice, P = 0.005). We noted that a fraction of oligodendrocytes located within the tracer-labeled clouds was not filled with biocytin (Fig. 3). Quantification revealed that 29% (26.4 ± 4.8 out of 90 ± 15.7 cells) of the PLP-GFP-positive cells within the barreloid volumes occupied by the biocytin-labeled networks were not connected to these networks. Thus, juvenile barreloidal coupling networks consist of astrocytes and oligodendrocytes, but there are also oligodendrocytes that apparently lack coupling. Barreloid Borders Are Mainly Formed by Weakly Coupled Oligodendrocytes The density of cell somata in the barreloids is higher at the borders than within the center (Van Der 1976) but it was still unclear how the different cell types are distributed within the barreloids, and which cells form its borders. Using PLP-GFP mice, we noted that oligodendrocytes were preferentially located along the barreloid borders. For quantitative evaluation co-labeling of nuclei and line scanning perpendicularly to the borders were performed, revealing that GFP-positive oligodendrocytes actually formed the border structures (Fig. 4A). To investigate the distribution of neurons and their processes, we stained slices for NeuN (Fig. 4B) and the axonal marker SMI (Fig. 4C), which indicated that neuronal somata were homogenously distributed while axons run essentially parallel to the barreloid borders. Next we analyzed the complete morphology and orientation of neurons within the barreloids. Therefore, we filled several neurons with the dye Alexa Fluor 594 by electroporation (5 slices). Image stacks of groups of fluorescent neurons were acquired and 3D reconstructions were performed to compute iso-surfaces (Fig. 5A,B). Calculated volume fragments were fitted into ellipsoids, a sum vector was calculated from the 3D orientation of the ellipsoids and after projection into a x-y plane the angle between sum vector and the barreloid borders was determined for each slice. The median of the angle between the neuronal processes and the border of the barreloids amounted to 6° (2°–39°, 5 slices from 5 mice) (Fig. 5C), similar to the parallel arrangement of axons and barreloid borders observed with SMI staining (Fig. 4C). Figure 4. View largeDownload slide Distinct distribution of oligodendrocytes and neurons within barreloid fields. Slices containing barreloids were prepared from juvenile PLP-GFP mice. (A) The distribution of oligodendrocytes within barreloid fields was revealed by staining for GFP (A1) while Hoechst staining was used to visualize barreloid borders (indicated by short gray lines in A2, B2, C2). The merged image (A3) demonstrates that barreloid borders are mainly composed of oligodendrocytes. (B) NeuN staining indicates that neuronal somata are evenly distributed across barreloid fields. (C) Staining for SMI demonstrates that axons run within the barreloid, mostly in parallel to the field borders. Scale bar 50 µm. (D) To quantitatively assess the distribution of cell types, horizontal line scans (distance 35 µm) across a barreloid border were performed and averaged. The graphs show mean changes in fluorescence intensities as arbitrary units (AU) ± SEM (blue lines, Hoechst; green line, GFP; red line, NeuN; yellow line, SMI), confirming that the borders were mainly formed by oligodendrocytes (D1, 11 slices from 3 mice). Neuronal cell bodies were evenly distributed (D2, 11 slices from 3 mice), and their axons mainly ran within the barreloid fields (D3, 13 slices from 3 mice). Figure 4. View largeDownload slide Distinct distribution of oligodendrocytes and neurons within barreloid fields. Slices containing barreloids were prepared from juvenile PLP-GFP mice. (A) The distribution of oligodendrocytes within barreloid fields was revealed by staining for GFP (A1) while Hoechst staining was used to visualize barreloid borders (indicated by short gray lines in A2, B2, C2). The merged image (A3) demonstrates that barreloid borders are mainly composed of oligodendrocytes. (B) NeuN staining indicates that neuronal somata are evenly distributed across barreloid fields. (C) Staining for SMI demonstrates that axons run within the barreloid, mostly in parallel to the field borders. Scale bar 50 µm. (D) To quantitatively assess the distribution of cell types, horizontal line scans (distance 35 µm) across a barreloid border were performed and averaged. The graphs show mean changes in fluorescence intensities as arbitrary units (AU) ± SEM (blue lines, Hoechst; green line, GFP; red line, NeuN; yellow line, SMI), confirming that the borders were mainly formed by oligodendrocytes (D1, 11 slices from 3 mice). Neuronal cell bodies were evenly distributed (D2, 11 slices from 3 mice), and their axons mainly ran within the barreloid fields (D3, 13 slices from 3 mice). Figure 5. View largeDownload slide Neuronal processes are aligned in parallel to the barreloid axis. (A) Neuronal morphology was assessed by electroporation of somata with Alexa Fluor 594 (red). (B) A 2-photon image stack has been used for 3D reconstruction and was successively overlaid with computed iso-surfaces (white). Individual iso-surfaces were used for an ellipsoid fitting routine. (C) Barreloid borders were labeled (dashed lines) in an aligned maximum intensity projection onto the x-y plane. Principal axes of ellipsoids fitted from (B) were averaged to generate a sum vector (arrow), which was used to determine the angle between barreloid orientation and neuronal processes (2° in the present example). Scale of the 3D grid units = 20 µm (A, B), 2D grid units = 50 µm (C). Figure 5. View largeDownload slide Neuronal processes are aligned in parallel to the barreloid axis. (A) Neuronal morphology was assessed by electroporation of somata with Alexa Fluor 594 (red). (B) A 2-photon image stack has been used for 3D reconstruction and was successively overlaid with computed iso-surfaces (white). Individual iso-surfaces were used for an ellipsoid fitting routine. (C) Barreloid borders were labeled (dashed lines) in an aligned maximum intensity projection onto the x-y plane. Principal axes of ellipsoids fitted from (B) were averaged to generate a sum vector (arrow), which was used to determine the angle between barreloid orientation and neuronal processes (2° in the present example). Scale of the 3D grid units = 20 µm (A, B), 2D grid units = 50 µm (C). The observation that barreloid borders were preferentially formed by oligodendrocytes led us to ask whether those oligodendrocytes were part of the panglial network. Using PLP-GFP mice, oligodendrocytes located on the barreloid border were selected and filled with biocytin during patch clamp recording (20 min). Two of 8 biocytin-filled oligodendrocytes (2 slices from 2 mice) were completely uncoupled (Fig. 6). The remaining 6 cells (6 slices from 6 mice) showed significantly reduced spread of biocytin (to 39.8 ± 4.8 cells, thereof 51.5% oligodendrocytes, and 48.5% astrocytes), as compared with filling oligodendrocytes in extra-barreloidal areas (network size 86.7 ± 13.9 cells, thereof 66.5 ± 8.4% oligodendrocytes, see Griemsmann et al. 2015), or with intrabarreloidal networks after astrocyte filling (cf. above). Again, as observed when filling astrocytes (Fig. 3), many (39%, i.e., 13.3 ± 2.2 out of 33.7 ± 3.8 cells) of the PLP-GFP-positive cells located within the tracer-filled networks were not biocytin-positive. Figure 6. View largeDownload slide Uncoupled oligodendrocytes are localized on barreloid borders. Slices containing barreloids were prepared from juvenile PLP-GFP mice. One oligodendrocyte on a barreloid border was identified by its intrinsic fluorescence and filled with Texas Red Dextran and biocytin. (A) The image shows the Hoechst staining (blue), the intrinsic GFP fluorescence (green) and the patched oligodendrocyte (Texas Red Dextran, pink). The gray lines indicate the barreloid border. (B) The biocytin channel (red) of the same cell and the merged picture (C) show that the patched oligodendrocyte is uncoupled. Scale bar 50 µm. The patched cell was initially located directly on the barreloid border. It shifted a few µm to the right due to technical issue regarding the patch technique. Scale bar, 25 µm. Figure 6. View largeDownload slide Uncoupled oligodendrocytes are localized on barreloid borders. Slices containing barreloids were prepared from juvenile PLP-GFP mice. One oligodendrocyte on a barreloid border was identified by its intrinsic fluorescence and filled with Texas Red Dextran and biocytin. (A) The image shows the Hoechst staining (blue), the intrinsic GFP fluorescence (green) and the patched oligodendrocyte (Texas Red Dextran, pink). The gray lines indicate the barreloid border. (B) The biocytin channel (red) of the same cell and the merged picture (C) show that the patched oligodendrocyte is uncoupled. Scale bar 50 µm. The patched cell was initially located directly on the barreloid border. It shifted a few µm to the right due to technical issue regarding the patch technique. Scale bar, 25 µm. Neuronal Activity Shapes Glial Coupling in Thalamic Barreloids Next we addressed in C57/Bl6J/N mice the question whether glial coupling in thalamic barreloids is controlled by neuronal activity (Fig. 7). After preincubation of thalamic slices from wild-type mice with TTX (0.5 µM) and ω-conotoxin GVIA (0.5 µM) for 3–4 h, individual astrocytes within the barreloid structures were filled (20 min) with a solution containing Texas Red Dextran, biocytin, and the fluorescent metabolizable glucose analog, 2-NBDG. Astrocytes were identified by incubating the tissue with SR101 prior to the recording. Under control conditions, that is, the same preincubation time in ACSF, 2-NBDG spread into 50.3 ± 1.7 neighboring cells (5 slices from 4 mice) while coupling was significantly reduced to 26.6 ± 1.0 cells after inhibition of neuronal activity (5 slices from 4 mice). Neuronal inhibition also entailed spherical network shapes (y/x-ratio 0.96 ± 0.02). After fixation of the tissue slices, we also compared the spread of biocytin. The number of biocytin-labeled cells after inhibition of neuronal activity was also significantly lower (54.1 ± 3.5 cells; 5 slices from 4 mice) as compared with control slices incubated for the same time in ACSF (88.2 ± 4.6 cells), and the biocytin-filled networks in TTX-containing solution were also spherical (y/x-ratio 0.99 ± 0.06; 5 slices from 4 mice). Figure 7. View largeDownload slide Neuronal activity modulates glial coupling in thalamic barreloids. Slices containing barreloids were prepared from juvenile C57Bl/6 mice. (A) An individual astrocyte within the barreloid was filled with biocytin (red, A1), Texas Red Dextran and the glucose analog, 2-NBDG (yellow, A2) after incubation for 3–4 h in ACSF. (B) Slices were incubated with TTX (0.5 µM) and ω-conotoxin GVIA (0.5 µM) for 3–4 h and an astrocyte was filled similar as in A. Scale bar 100 µm. (C) Inhibition of neuronal activity decreased the spread of 2-NBDG and biocytin. The data are an average from 5 slices from 4 mice for both, control and TTX+ω-conotoxin. Number of mice is given in bar graphs. Figure 7. View largeDownload slide Neuronal activity modulates glial coupling in thalamic barreloids. Slices containing barreloids were prepared from juvenile C57Bl/6 mice. (A) An individual astrocyte within the barreloid was filled with biocytin (red, A1), Texas Red Dextran and the glucose analog, 2-NBDG (yellow, A2) after incubation for 3–4 h in ACSF. (B) Slices were incubated with TTX (0.5 µM) and ω-conotoxin GVIA (0.5 µM) for 3–4 h and an astrocyte was filled similar as in A. Scale bar 100 µm. (C) Inhibition of neuronal activity decreased the spread of 2-NBDG and biocytin. The data are an average from 5 slices from 4 mice for both, control and TTX+ω-conotoxin. Number of mice is given in bar graphs. Two-photon online analysis was used to test whether there were also uncoupled astrocytes within the volume spanned by the tracer-filled network. Slices were obtained from PLP-GFP mice, labeled with SR101 and individual astrocytes were filled with 2-NBDG for 20 min. The strong fluorescence intensity of the 2-NBDG-filled patch pipette, which covered part of the region of interest, hampered reliable online analysis of the complete coupled network. Nevertheless, there were obviously many uncoupled, that is, 2-NBDG-negative/SR101-positive astrocytes, which apart from the region obscured by the pipette fluorescence amounted to about 40% of the SR101-positive astrocytes within that area, both under normal conditions, and after adding TTX (5 slices from 4 mice). Thus, within the barreloids many astrocytes and oligodendrocytes were not part of the coupling network. Discussion It is now increasingly appreciated that astrocytes comprise a heterogeneous cell population, and possess distinct properties to meet the respective needs of the various brain regions and microcircuits (Matyash and Kettenmann 2010). Compared with other areas of the CNS, little is known about glial cells in the thalamus, but recent studies revealed that their molecular and functional features differ significantly from those in the hippocampus or cortex (Höft et al. 2014; Griemsmann et al. 2015). In this study, we have characterized cells located in the barreloids of the somatosensory thalamus, which represent cell clusters receiving sensory input from individual vibrissae and transmitting their output to the corresponding barrels in the neocortex (Land et al. 1995). We report here that gap junction networks in the thalamus comprise astrocytes and oligodendrocytes. Network size is limited by barreloid borders, which are mainly formed by weakly coupled oligodendrocytes and neuronal activity affects the network size and shape. After defining an appropriate cutting angle, we were able to visualize barreloid structures in acute slices prepared from the ventrobasal nucleus of the thalamus. It has been reported previously that the shape and extent of astroglial gap junction networks can be tightly linked to neuronal functional units, for example, in case of the barrels in the somatosensory cortex (Houades et al. 2008), olfactory glomeruli (Roux et al. 2011) or along the plane of the dendritic tree of Purkinje cells in the cerebellum (Müller et al. 1996). We report here that in the juvenile ventrobasal thalamus, the barreloid borders exert coupling barriers. In contrast to cortical barrels and olfactory glomeruli, but similar to extra-barreloidal areas of the adult thalamus (Griemsmann et al. 2015), networks in the barreloids not only consisted of coupled astrocytes but equally of astrocytes and oligodendrocytes. Our immunostainings did not indicate septal regions with decreased density of neuronal somata as it has been reported earlier (Van Der 1976; Land et al. 1995). However, analysis of PLP-GFP mice revealed that the borders between barreloids are mainly formed by oligodendrocytes, which showed comparatively weak coupling, explaining why the tracer-filled networks within barreloids were asymmetrical and did not extend towards neighboring barreloids. We noted that many PLP-GFP-positive cells and also some astrocytes located within the volume spanned by the tracer-labeled networks were biocytin-negative. Uncoupled astrocytes have also been observed in the juvenile hippocampus (Houades et al. 2006), which, however, was due to the immature functional stage of these cells (Wallraff et al. 2004; Schools et al. 2006; Strohschein, 2011). In contrast, the presence of uncoupled cells we report here cannot be attributed to the early age because in the thalamus tracer coupling reaches a steady state already by the second postnatal week (Griemsmann et al. 2015). Although it remains to be shown whether these tracer-negative glial cells were really uncoupled or rather part of another network, our data provide further evidence for functional heterogeneity among thalamic glial cells. Astrocytes in the ventrobasal thalamus differ with respect to the expression of AMPA/KA type of glutamate receptors (Höft et al. 2014) and other studies have shown differences in their responsiveness to stimulation of sensory versus corticothalamic afferences (Parri et al. 2010; Pirttimaki and Parri 2012). It remains to be shown whether coupled or uncoupled astrocytes have distinct functional properties and communicate differently with barreloidal neurons. It has been shown that neuronal activity may shape astrocyte coupling networks. For example, increased coupling between astrocytes has been observed in co-cultures with neurons (Fischer and Kettenmann 1985; Rouach et al. 2000) or upon stimulation in the optic nerve (Marrero and Orkand 1996) and altered phosphorylation of Cx43 has been proposed as an mediator of this activity-dependent effect (reviewed by Giaume et al. 2010). However, in the hippocampus, a region where coupling critically depends on Cx43 (Griemsmann et al. 2015), neuronal activity regulated the spread of energy metabolites (including 2-NBDG) through the network, but did not alter gap junction channel permeability as revealed by tracer filling (Rouach et al. 2008). As opposed to the situation in the hippocampus, inhibition of neuronal activity in the olfactory bulb did decrease astroglial gap junction permeability (Roux et al. 2011). Utilizing connexin knockout mice and considering the delayed onset of Cx30 expression, the latter study suggested that in olfactory glomeruli, Cx30 is the target of activity-dependence of astrocytic coupling. We show here that suppression of neuronal activity significantly, and to a similar proportion, reduces the spread of biocytin and 2-NBDG between glial cells in thalamic barreloids. Our data are thus in line with the findings by Roux et al. (2011), given the predominating role of Cx30 for coupling in the thalamus (Griemsmann et al. 2015). Based on results of the latter study and the activity-dependence of coupling shown here, we propose that panglial coupling in the barreloids of the juvenile thalamus is mainly mediated by heterotypic Cx30:Cx32 channels. What might be the mechanistic link between neuronal activity and gap junction coupling in the thalamus? It has been proposed that fluctuations in extracellular K+ concentration may control Cx30-formed gap junctions in a Kir channel-dependent manner (Roux et al. 2011). Interestingly, 2 functionally distinct subpopulations of thalamic astrocytes have been described, expressing or lacking AMPA receptors and differing in Kir current densities (Höft et al. 2014). It is thus tempting to speculate that some of the AMPA receptor bearing astrocytes belonged to those remaining tracer-negative in the coupling experiments. Astrocytes are in the strategic position to take up glucose from the blood through their end-feet covering the capillaries (Belanger et al. 2011), and functional gap junctions of astrocytes are critically required for energetic supply to maintain synaptic activity (Rouach et al. 2008). What is the specific role of oligodendrocytes, which made up 60% of coupled cells in barreloids? Our data show that upon filling an astrocyte with 2-NBDG, the fluorescent glucose spreads into oligodendrocytes, and along myelin sheaths. Given that oligodendrocytes are metabolically highly active cells (Amaral et al. 2016), our findings suggest that astrocyte-oligodendrocyte coupling may be important for metabolic support of myelinated axons (Fünfschilling et al. 2012; Lee et al. 2012; Morrison et al. 2013). Here, coupling might serve as a back-up pathway to fuel the respective metabolic pathways, for example, lipid synthesis in oligodendrocytes and myelination. What might be the functional significance of isolating barreloidal glial networks from each other? In case of neocortical barrels, it has been proposed that glial network separation is required to narrow down the somatotopic whisker signaling pathway (Houades et al. 2008), a concept that similarly might apply to barreloidal coupling. There, the implicit assumption is that glial networks would otherwise synchronize larger neuronal assemblies or distribute information unspecifically among these. In the thalamic barreloids, little is known about the ability of glial networks to sense and modulate neuronal behavior. It is nonetheless tempting to speculate that isolation of glial networks limits output-relevant glia-neuron interactions to the barreloid itself. Along the same line of thought, spatially restricted glial networks could also help stabilizing somatotopic projection by preventing generalization of aberrant neuronal activity. As a metabolic implication one may envision that the restricted barreloidal networks lead to uncoupling in terms of energy supply, thereby limiting the metabolic load to the respectively active vibrissae signaling pathways. Indeed, deprivation experiments have revealed such a pathway-specificity of neuronal metabolism (Land and Akhtar 1987). Future studies will have to unravel the specific role of the different glial cell types composing panglial networks in maintaining and fine tuning synaptic and axonal activity in barreloids under normal conditions and in the diseased brain. Supplementary Material Supplementary data is available at Cerebral Cortex online. Author Contributions L.C., C.P., S.G., R.J., C.H., and C.S. designed the research. L.C., C.P., R.J., C.H., H.K., and C.S. wrote the manuscript. L.C., C.P., R.J., A.T., C.H., and C.S. conducted experiments and analyzed data. Funding German Research Foundation (SPP1757: STE 552/4 to C.S., SFB1089: B03 to C.H., KE 329/28 to H.K., and SPP1757 to C.S., C.H., and H.K.), NRW-Rückkehrerprogramm (to C.H.), and Human Frontiers Science Program (HFSP: RGY-0084/2012, to C.H.). Notes L.C. was a fellow of the BONFOR program of Bonn Medical Faculty. We thank Christian Giaume, Paris for helpful discussions and Dilaware Khan, Bonn for technical support. Conflict of Interest: Authors declare no competing financial interests. References Amaral AI, Hadera MG, Tavares JM, Kotter MR, Sonnewald U. 2016. Characterization of glucose-related metabolic pathways in differentiated rat oligodendrocyte lineage cells. Glia . 64: 21– 34. Google Scholar CrossRef Search ADS PubMed  Anders S, Minge D, Griemsmann S, Herde MK, Steinhäuser C, Henneberger C. 2014. Spatial properties of astrocyte gap junction coupling in the rat hippocampus. Philos Trans R Soc Lond B Biol Sci . 369: 20120600. Google Scholar CrossRef Search ADS   Bedner P, Steinhäuser C, Theis M. 2012. Functional redundancy and compensation among members of gap junction protein families? Biochim Biophys Acta . 1818: 1971– 1984. Google Scholar CrossRef Search ADS PubMed  Belanger M, Allaman I, Magistretti PJ. 2011. Brain energy metabolism: focus on astrocyte-neuron metabolic cooperation. Cell Metab . 14: 724– 738. Google Scholar CrossRef Search ADS PubMed  Fischer G, Kettenmann H. 1985. Cultured astrocytes form a syncytium after maturation. Exp Cell Res . 159: 273– 279. Google Scholar CrossRef Search ADS PubMed  Fünfschilling U, Supplie LM, Mahad D, Boretius S, Saab AS, Edgar J, Brinkmann BG, Kassmann CM, Tzvetanova ID, Mobius W, et al.  . 2012. Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. Nature . 485: 517– 521. Google Scholar PubMed  Fuss B, Mallon B, Phan T, Ohlemeyer C, Kirchhoff F, Nishiyama A, Macklin WB. 2000. Purification and analysis of in vivo-differentiated oligodendrocytes expressing the green fluorescent protein. Dev Biol . 218: 259– 274. Google Scholar CrossRef Search ADS PubMed  Giaume C, Koulakoff A, Roux L, Holcman D, Rouach N. 2010. Astroglial networks: a step further in neuroglial and gliovascular interactions. Nat Rev Neurosci . 11: 87– 99. Google Scholar CrossRef Search ADS PubMed  Griemsmann S, Höft SP, Bedner P, Zhang J, von SE, Beinhauer A, Degen J, Dublin P, Cope DW, Richter N, et al.  . 2015. Characterization of panglial gap junction networks in the thalamus, neocortex, and hippocampus reveals a unique population of glial cells. Cereb Cortex . 25: 3420– 3433. Google Scholar CrossRef Search ADS PubMed  Höft S, Griemsmann S, Seifert G, Steinhäuser C. 2014. Heterogeneity in expression of functional ionotropic glutamate and GABA receptors in astrocytes across brain regions: insights from the thalamus. Philos Trans R Soc Lond B Biol Sci . 369: 20130602. Google Scholar CrossRef Search ADS PubMed  Houades V, Kirchhoff F, Ezan P, Rouach N, Koulakoff A, Giaume C. 2006. Shapes of astrocyte networks in the juvenile brain. Neuron Glia Biol . 2: 3– 14. Google Scholar CrossRef Search ADS PubMed  Houades V, Koulakoff A, Ezan P, Seif I, Giaume C. 2008. Gap junction-mediated astrocytic networks in the mouse barrel cortex. J Neurosci . 28: 5207– 5217. Google Scholar CrossRef Search ADS PubMed  Jones EG, editor. 2007. The thalamus . 2nd ed. New York: Cambridge University Press. Land PW, Akhtar ND. 1987. Chronic sensory deprivation affects cytochrome oxidase staining and glutamic acid decarboxylase immunoreactivity in adult rat ventrobasal thalamus. Brain Res . 425: 178– 181. Google Scholar CrossRef Search ADS PubMed  Land PW, Buffer SA Jr, Yaskosky JD. 1995. Barreloids in adult rat thalamus: three-dimensional architecture and relationship to somatosensory cortical barrels. J Comp Neurol . 355: 573– 588. Google Scholar CrossRef Search ADS PubMed  Lee Y, Morrison BM, Li Y, Lengacher S, Farah MH, Hoffman PN, Liu Y, Tsingalia A, Jin L, Zhang PW, et al.  . 2012. Oligodendroglia metabolically support axons and contribute to neurodegeneration. Nature . 487: 443– 448. Google Scholar CrossRef Search ADS PubMed  Marrero H, Orkand RK. 1996. Nerve impulses increase glial intercellular permeability. Glia . 16: 285– 289. Google Scholar CrossRef Search ADS PubMed  Matthias K, Kirchhoff F, Seifert G, Hüttmann K, Matyash M, Kettenmann H, Steinhäuser C. 2003. Segregated expression of AMPA-type glutamate receptors and glutamate transporters defines distinct astrocyte populations in the mouse hippocampus. J Neurosci . 23: 1750– 1758. Google Scholar PubMed  Matyash V, Kettenmann H. 2010. Heterogeneity in astrocyte morphology and physiology. Brain Res Rev . 63: 2– 10. Google Scholar CrossRef Search ADS PubMed  Morrison BM, Lee Y, Rothstein JD. 2013. Oligodendroglia: metabolic supporters of axons. Trends Cell Biol . 23: 644– 651. Google Scholar CrossRef Search ADS PubMed  Mosconi T, Woolsey TA, Jacquin MF. 2010. Passive vs. active touch-induced activity in the developing whisker pathway. Eur J Neurosci . 32: 1354– 1363. Google Scholar CrossRef Search ADS PubMed  Müller T, Möller T, Neuhaus J, Kettenmann H. 1996. Electrical coupling among Bergmann glial cells and its modulation by glutamate receptor activation. Glia . 17: 274– 284. Google Scholar CrossRef Search ADS PubMed  Nevian T, Helmchen F. 2007. Calcium indicator loading of neurons using single-cell electroporation. Pflugers Arch . 454: 675– 688. Google Scholar CrossRef Search ADS PubMed  Nolte C, Matyash M, Pivneva T, Schipke CG, Ohlemeyer C, Hanisch UK, Kirchhoff F, Kettenmann H. 2001. GFAP promoter-controlled EGFP-expressing transgenic mice: a tool to visualize astrocytes and astrogliosis in living brain tissue. Glia . 33: 72– 86. Google Scholar CrossRef Search ADS PubMed  Parri HR, Gould TM, Crunelli V. 2010. Sensory and cortical activation of distinct glial cell subtypes in the somatosensory thalamus of young rats. Eur J Neurosci . 32: 29– 40. Google Scholar CrossRef Search ADS PubMed  Pirttimaki TM, Parri HR. 2012. Glutamatergic input-output properties of thalamic astrocytes. Neuroscience . 205: 18– 28. Google Scholar CrossRef Search ADS PubMed  Rouach N, Glowinski J, Giaume C. 2000. Activity-dependent neuronal control of gap-junctional communication in astrocytes. J Cell Biol . 149: 1513– 1526. Google Scholar CrossRef Search ADS PubMed  Rouach N, Koulakoff A, Abudara V, Willecke K, Giaume C. 2008. Astroglial metabolic networks sustain hippocampal synaptic transmission. Science . 322: 1551– 1555. Google Scholar CrossRef Search ADS PubMed  Roux L, Benchenane K, Rothstein JD, Bonvento G, Giaume C. 2011. Plasticity of astroglial networks in olfactory glomeruli. Proc Natl Acad Sci U S A . 108: 18442– 18446. Google Scholar CrossRef Search ADS PubMed  Schools GP, Zhou M, Kimelberg HK. 2006. Development of gap junctions in hippocampal astrocytes: evidence that whole cell electrophysiological phenotype is an intrinsic property of the individual cell. J Neurophysiol . 96: 1383– 1392. Google Scholar CrossRef Search ADS PubMed  Sugitani M, Yano J, Sugai T, Ooyama H. 1990. Somatotopic organization and columnar structure of vibrissae representation in the rat ventrobasal complex. Exp Brain Res . 81: 346– 352. Google Scholar CrossRef Search ADS PubMed  Strohschein S. 2011. Einfluss des Aquaporin 4 Kanals auf die Kaliumpufferung und die Gap junction Kopplung im Hippocampus der Maus [dissertation]. University of Bonn, urn:nbn:de:hbz:5 n-30700. Theis M, Giaume C. 2012. Connexin-based intercellular communication and astrocyte heterogeneity. Brain Res . 1487: 88– 98. Google Scholar CrossRef Search ADS PubMed  Van Der LH. 1976. Barreloids in mouse somatosensory thalamus. Neurosci Lett . 2: 1– 6. Google Scholar CrossRef Search ADS PubMed  Wallraff A, Odermatt B, Willecke K, Steinhäuser C. 2004. Distinct types of astroglial cells in the hippocampus differ in gap junction coupling. Glia . 48: 36– 43. Google Scholar CrossRef Search ADS PubMed  Wong-Riley M. 1979. Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Res . 171: 11– 28. Google Scholar CrossRef Search ADS PubMed  © The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
TI - Barreloid Borders and Neuronal Activity Shape Panglial Gap Junction-Coupled Networks in the Mouse Thalamus
JF - Cerebral Cortex
DO - 10.1093/cercor/bhw368
DA - 2018-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/barreloid-borders-and-neuronal-activity-shape-panglial-gap-junction-bsmJanaJDK
SP - 213
EP - 222
VL - 28
IS - 1
DP - DeepDyve
ER -